Smartphone-Based Activity Recognition Using Multistream Movelets Combining Accelerometer and Gyroscope Data.

Publisher: MDPI Country of Publication: Switzerland NLM ID: 101204366 Publication Model: Electronic Cited Medium: Internet ISSN: 1424-8220 (Electronic) Linking ISSN: 14248220 NLM ISO Abbreviation: Sensors (Basel) Subsets: MEDLINE

Original Publication: Basel, Switzerland : MDPI, c2000-

Exercise* ; Smartphone*; Accelerometry/methods ; Humans ; Sitting Position ; Standing Position

Physical activity patterns can reveal information about one's health status. Built-in sensors in a smartphone, in comparison to a patient's self-report, can collect activity recognition data more objectively, unobtrusively, and continuously. A variety of data analysis approaches have been proposed in the literature. In this study, we applied the movelet method to classify the activities performed using smartphone accelerometer and gyroscope data, which measure a phone's acceleration and angular velocity, respectively. The movelet method constructs a personalized dictionary for each participant using training data and classifies activities in new data with the dictionary. Our results show that this method has the advantages of being interpretable and transparent. A unique aspect of our movelet application involves extracting unique information, optimally, from multiple sensors. In comparison to single-sensor applications, our approach jointly incorporates the accelerometer and gyroscope sensors with the movelet method. Our findings show that combining data from the two sensors can result in more accurate activity recognition than using each sensor alone. In particular, the joint-sensor method reduces errors of the gyroscope-only method in differentiating between standing and sitting. It also reduces errors in the accelerometer-only method when classifying vigorous activities.

Res Q Exerc Sport. 2000 Jun;71 Suppl 2:1-14. (PMID: 25680007)
J Neuroeng Rehabil. 2016 Jan 20;13:5. (PMID: 26792670)
JAMA Surg. 2020 Feb 1;155(2):123-129. (PMID: 31657854)
J Acad Nutr Diet. 2014 Feb;114(2):199-208. (PMID: 24290836)
Front Neurol. 2012 Nov 07;3:158. (PMID: 23162528)
JMIR Ment Health. 2016 May 05;3(2):e16. (PMID: 27150677)
NPJ Digit Med. 2021 Oct 18;4(1):148. (PMID: 34663863)
Sensors (Basel). 2015 Dec 04;15(12):30636-52. (PMID: 26690163)
Electron J Stat. 2012;6:559-578. (PMID: 23293708)
Biostatistics. 2021 Apr 10;22(2):331-347. (PMID: 31545345)
JMIR Mhealth Uhealth. 2019 Aug 23;7(8):e12649. (PMID: 31444874)
Sensors (Basel). 2014 Jun 10;14(6):10146-76. (PMID: 24919015)
Sensors (Basel). 2020 Jul 02;20(13):. (PMID: 32630752)
Med Sci Sports Exerc. 2014 Sep;46(9):1859-66. (PMID: 25134005)
IEEE Access. 2020;8:210816-210836. (PMID: 33344100)
Am J Epidemiol. 2005 Feb 15;161(4):389-98. (PMID: 15692083)
Br J Sports Med. 2003 Jun;37(3):197-206; discussion 206. (PMID: 12782543)
Sensors (Basel). 2020 Apr 14;20(8):. (PMID: 32295298)
Physiol Meas. 2014 Nov;35(11):2269-86. (PMID: 25340659)
Neuropsychopharmacology. 2021 Jan;46(1):45-54. (PMID: 32679583)

U01 HL145386 United States HL NHLBI NIH HHS; U01HL145386 National Heart Lung and Blood Institute

Keywords: accelerometer; activity recognition; digital phenotyping; gyroscope; movelet; sensor; smartphone

Date Created: 20220412 Date Completed: 20220413 Latest Revision: 20240826

20240826

PMC9002497

10.3390/s22072618

35408232